This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TMDSEVM6678: Hyperlink code example can be executed only once

Part Number: TMDSEVM6678

Hi,

I am testing the TMDSEVM6678LE by running the supplied code examples. I am using only one core out of 8 that the c6678 has. The problem is: when I execute the hyperlink example, it runs only once. To re-run the code I need to unplug & plug the board. Nothing else helps, Terminate & CPU buttons become useless.

1) What do I have to do in-order to re-run the code?

2) Can I reset one core and keep all the other running?

3) Where can I find a 'timeout' example routine?

4) If the current working core gets stuck, is it possible to trigger another core?

Kind regards

Moti

  • Hi Moti,

    Which RTOS SDK are you using?

    Best Regards,
    Yordan
  • c667x_4_01_00_06
  • Hi Yordan,
    I'm using RTOS c667x_4_01_00_06. Does it help you?
    One more question:
    I'm using only core 0 right now and I would like to add more cores. How do I do that?

    Thanks,
    Moti
  • Hi Moti,

    Yes, I've notified the RTOS team. They will post their feedback directly here.

    Best Regards,
    Yordan
  • Hi,

    1) Are you testing the memory mapped example? One side of Hyperlink is TI 6678 EVM, what is the other side of the connection? The driver examples in general are not designed to re-run. It typically runs from a fresh powered on system, with GEL file to initialize the board. Some example may have code to release the resource and can re-run without power cycle the board.

    For the Hyperlink, I saw there is hyplnkReset() at the end of the mainb routine, where the Hyperlink PSC domain is powered down. So it should be able to rerun. If not, you may need to step the code where it stuck.

    2) yes

    3) Where can I find a 'timeout' example routine? ====> please explain what you are looking for?

    4) If the current working core gets stuck, is it possible to trigger another core?====> the stuck core can't trigger another core. You may have a master core to keep track each core status, if one core stuck, you trigger another?

    Regards, Eric
  • Hi Yordan & Eric,

    Thank you for answers so far.

    As you already know, I'm currently using only one core that gets stuck at the infraDMA part of the hyperlinkExample. For better understanding, I want to add more cores to my activated example but I can't find the way.

    So:

    1. How can I add more cores to my example?

    2. How can setup particular cores to send data to other particular cores?

    3. About the 'timeout' example, I'm looking for a code section that reads current clock, waits for some time(as long as I wish) till the QMSS set a request otherwise go on. Clear enough?

    Regards,

    Moty

  • Hi,

    The driver example is a single core test. You may try to run it on cores other than core 0. You still need to run the GEL to initialize the board. CPPI, QMSS initialization may only runs on core 0, please double check code and modify it to run from other cores.

    Your question is not related to Hyperlink, you may look at the QMSS and CPPI driver if any examples available.

    Regards, Eric